Mercurial > hg > aimc

changeset 22:645cfd371cff

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
- Added scripts directory with a few basic scripts for testing modules and interfacting with matlab
author tomwalters
date Mon, 22 Feb 2010 18:17:14 +0000
parents d386805133f1
children 491b1b1d1dc5
files src/Scripts/Features_test.py src/Scripts/Profiles_test.py src/Scripts/aa161p119s100t.wav src/Scripts/profile_features.cfg
diffstat 4 files changed, 226 insertions(+), 0 deletions(-) [+]
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/src/Scripts/Features_test.py	Mon Feb 22 18:17:14 2010 +0000
@@ -0,0 +1,110 @@
+#!/usr/bin/env python
+# encoding: utf-8
+#
+# AIM-C: A C++ implementation of the Auditory Image Model
+# http://www.acousticscale.org/AIMC
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+"""
+Profiles_test.py
+
+Created by Thomas Walters on 2010-02-22.
+Copyright 2010 Thomas Walters <tom@acousticscale.org>
+Test the AIM-C model from filterbank to SSI profiles
+"""
+
+import aimc
+from scipy.io import wavfile
+from scipy import io
+import scipy
+import pylab
+from itertools import izip, chain, repeat
+
+def grouper(n, iterable, padvalue=None):
+    "grouper(3, 'abcdefg', 'x') --> ('a','b','c'), ('d','e','f'), ('g','x','x')"
+    return izip(*[chain(iterable, repeat(padvalue, n-1))]*n)
+
+def main():
+  wave_path = "/Users/Tom/Documents/Work/PhD/HTK-AIM/Sounds/"
+  features_path = "/Users/Tom/Documents/Work/PhD/HTK-AIM/work08-jess-original-rec_rubber/features/"
+  
+  file_name = "aa/aa161.1p119.4s100.0t+000itd"
+  
+  wave_suffix = ".wav"
+  features_suffix = ".mat"
+  
+  frame_period_ms = 10;
+    
+  wave_filename = wave_path + file_name + wave_suffix
+  features_filename = features_path + file_name + features_suffix
+  
+  (sample_rate, input_wave) = wavfile.read(wave_filename)
+  wave_length = input_wave.size
+  buffer_length = int(frame_period_ms * sample_rate / 1000)
+ 
+  #pylab.plot(input_wave)
+  #pylab.show()
+  
+  input_sig = aimc.SignalBank()
+  input_sig.Initialize(1, buffer_length, sample_rate)
+  parameters = aimc.Parameters()
+  parameters.Load("src/Scripts/profile_features.cfg")
+  mod_gt = aimc.ModuleGammatone(parameters)
+  mod_hl = aimc.ModuleHCL(parameters)
+  mod_profile = aimc.ModuleSlice(parameters)
+  mod_scaler = aimc.ModuleScaler(parameters)
+  mod_features = aimc.ModuleGaussians(parameters)
+  mod_gt.AddTarget(mod_hl)
+  mod_hl.AddTarget(mod_profile)
+  mod_profile.AddTarget(mod_scaler)
+  mod_scaler.AddTarget(mod_features)
+  mod_gt.Initialize(input_sig)
+  
+  correct_count = 0;
+  incorrect_count  = 0;
+  
+  scaled_wave = []
+  for sample in input_wave:
+    scaled_wave.append(float(sample / float(pow(2,15) - 1)))
+  i = 0
+  
+  wave_chunks = grouper(buffer_length, scaled_wave, 0)
+
+  out_frames = []
+  for chunk in wave_chunks:
+    i = 0
+    for sample in chunk:
+      input_sig.set_sample(0, i, float(sample))
+      i += 1
+    mod_gt.Process(input_sig)
+    out_sig = mod_features.GetOutputBank()
+    
+    channel_count = out_sig.channel_count()
+    out_buffer_length = out_sig.buffer_length()
+    cfs = scipy.zeros((channel_count))
+    out = scipy.zeros((channel_count, out_buffer_length))
+  
+    for ch in range(0, channel_count):
+      for i in range(0, out_buffer_length):  
+        out[ch, i] = out_sig.sample(ch, i)
+    out_frames.append(out)
+    
+  outmat = dict(profile_out=out_frames)
+  io.savemat("src/Scripts/features_out.mat", outmat)
+
+  pass
+
+
+if __name__ == '__main__':
+  main()
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/src/Scripts/Profiles_test.py	Mon Feb 22 18:17:14 2010 +0000
@@ -0,0 +1,108 @@
+#!/usr/bin/env python
+# encoding: utf-8
+#
+# AIM-C: A C++ implementation of the Auditory Image Model
+# http://www.acousticscale.org/AIMC
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+"""
+Profiles_test.py
+
+Created by Thomas Walters on 2010-02-22.
+Copyright 2010 Thomas Walters <tom@acousticscale.org>
+Test the AIM-C model from filterbank to SSI profiles
+"""
+
+import aimc
+from scipy.io import wavfile
+from scipy import io
+import scipy
+import pylab
+from itertools import izip, chain, repeat
+
+def grouper(n, iterable, padvalue=None):
+    "grouper(3, 'abcdefg', 'x') --> ('a','b','c'), ('d','e','f'), ('g','x','x')"
+    return izip(*[chain(iterable, repeat(padvalue, n-1))]*n)
+
+def main():
+  wave_path = "/Users/Tom/Documents/Work/PhD/HTK-AIM/Sounds/"
+  features_path = "/Users/Tom/Documents/Work/PhD/HTK-AIM/work08-jess-original-rec_rubber/features/"
+  
+  file_name = "aa/aa161.1p119.4s100.0t+000itd"
+  
+  wave_suffix = ".wav"
+  features_suffix = ".mat"
+  
+  frame_period_ms = 10;
+    
+  wave_filename = wave_path + file_name + wave_suffix
+  features_filename = features_path + file_name + features_suffix
+  
+  (sample_rate, input_wave) = wavfile.read(wave_filename)
+  wave_length = input_wave.size
+  buffer_length = int(frame_period_ms * sample_rate / 1000)
+ 
+  #pylab.plot(input_wave)
+  #pylab.show()
+  
+  input_sig = aimc.SignalBank()
+  input_sig.Initialize(1, buffer_length, sample_rate)
+  parameters = aimc.Parameters()
+  parameters.Load("src/Scripts/profile_features.cfg")
+  mod_gt = aimc.ModuleGammatone(parameters)
+  mod_hl = aimc.ModuleHCL(parameters)
+  mod_profile = aimc.ModuleSlice(parameters)
+  mod_scaler = aimc.ModuleScaler(parameters)
+  mod_gt.AddTarget(mod_hl)
+  mod_hl.AddTarget(mod_profile)
+  mod_profile.AddTarget(mod_scaler)
+  mod_gt.Initialize(input_sig)
+  
+  correct_count = 0;
+  incorrect_count  = 0;
+  
+  scaled_wave = []
+  for sample in input_wave:
+    scaled_wave.append(float(sample / float(pow(2,15) - 1)))
+  i = 0
+  
+  wave_chunks = grouper(buffer_length, scaled_wave, 0)
+
+  out_frames = []
+  for chunk in wave_chunks:
+    i = 0
+    for sample in chunk:
+      input_sig.set_sample(0, i, float(sample))
+      i += 1
+    mod_gt.Process(input_sig)
+    out_sig = mod_scaler.GetOutputBank()
+    
+    channel_count = out_sig.channel_count()
+    out_buffer_length = out_sig.buffer_length()
+    cfs = scipy.zeros((channel_count))
+    out = scipy.zeros((channel_count, out_buffer_length))
+  
+    for ch in range(0, channel_count):
+      for i in range(0, out_buffer_length):  
+        out[ch, i] = out_sig.sample(ch, i)
+    out_frames.append(out)
+    
+  outmat = dict(profile_out=out_frames)
+  io.savemat("src/Scripts/profile_out.mat", outmat)
+
+  pass
+
+
+if __name__ == '__main__':
+  main()
Binary file src/Scripts/aa161p119s100t.wav has changed
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/src/Scripts/profile_features.cfg	Mon Feb 22 18:17:14 2010 +0000
@@ -0,0 +1,8 @@
+gtfb.channel_count=200
+gtfb.min_frequency=86.0
+gtfb.max_frequency=16000.0
+nap.do_lowpass=true
+nap.lowpass_cutoff=100.0
+slice.temporal=false
+slice.all=true
+slice.normalize=true
\ No newline at end of file